基于XML属性值的PHP If / Then

Question

I have an XML file that I'm reading as a SimpleXML Element. In that file I have the following elements:

<files>
  <file>
    <type>GIF</type>
    <url>http://dl.server.com/1.GIF</url>
  </file>
  <file>
    <type>JPG</type>
    <url>http://dl.server.com/2.JPG</url>
  </file>
  <file>
    <type>TIF</type>
    <url>http://dl.server.com/1.TIF</url>
  </file>
  <file>
    <type>EPS</type>
    <url>http://dl.server.com/1.EPS</url>
  </file>
  <file>
    <type>LEPS</type>
    <url>http://dl.server.com/2.EPS</url>
  </file>
</files>

I was using a foreach() to loop through and based on the element's value, was performing an action. Now I need to look at all the files->file->type values and if "LEPS" exists, use that URL, otherwise if "EPS" exists, use that URL, and if neither exists, do nothing.

My struggle is with the XML node/element/property terms and not being able to find a way to query if files->file->type = "LEPS" is true or not. While I know how to check for attributes (isset(element['attributename']), I'm not sure how to check for an element with a specific property value.

Sorry for the elementary question!

Answer 1

You can use a predicate to check the <type> and those of the sibling file/types.

//files/file
[
    type='LEPS' or
    type='EPS' and not(
        following-sibling::file/type='LEPS' or
        preceding-sibling::file/type='LEPS'
    )
]/url

---

Example:

(This example uses the DOM extension rather than SimpleXML for reasons of taste. The xpath expression itself is agnostic and can be applied equally in either.)

/* $xml = your xml string */

$dom = new DOMDocument();
$dom->loadXML($xml);
$xpath = new DOMXPath($dom);

$query = <<<'XPATH'
//files/file
[
    type='LEPS' or
    type='EPS' and not(
        following-sibling::file/type='LEPS' or
        preceding-sibling::file/type='LEPS'
    )
]/url
XPATH;

foreach ($xpath->query($query) as $node) {
    echo $dom->saveXML($node), "\n";
}

Output:

<url>http://dl.server.com/2.EPS</url>

Answer 2

In basic pseudo, you could, in your foreach, check your type, if it matches, store the url for later use and break out of the loop.

Something similar to

url = null
foreach node in element:
    if node->type == 'LEPS':
        url = node->url
        break
    else if node->type == 'EPS':
        url = node->url

// if url is null, no result found